US9969507B2 - Method for performing diagnostics of a structure subject to loads and system for implementing said method - Google Patents

Method for performing diagnostics of a structure subject to loads and system for implementing said method Download PDF

Info

Publication number
US9969507B2
US9969507B2 US13/933,964 US201313933964A US9969507B2 US 9969507 B2 US9969507 B2 US 9969507B2 US 201313933964 A US201313933964 A US 201313933964A US 9969507 B2 US9969507 B2 US 9969507B2
Authority
US
United States
Prior art keywords
state
values
parameter
relevant
points
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active, expires
Application number
US13/933,964
Other languages
English (en)
Other versions
US20140012461A1 (en
Inventor
Michele Iannone
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Alenia Aermacchi SpA
Original Assignee
Alenia Aermacchi SpA
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Alenia Aermacchi SpA filed Critical Alenia Aermacchi SpA
Assigned to ALENIA AERMACCHI S.P.A. reassignment ALENIA AERMACCHI S.P.A. ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: IANNONE, MICHELE
Publication of US20140012461A1 publication Critical patent/US20140012461A1/en
Application granted granted Critical
Publication of US9969507B2 publication Critical patent/US9969507B2/en
Active legal-status Critical Current
Adjusted expiration legal-status Critical

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B64—AIRCRAFT; AVIATION; COSMONAUTICS
    • B64F—GROUND OR AIRCRAFT-CARRIER-DECK INSTALLATIONS SPECIALLY ADAPTED FOR USE IN CONNECTION WITH AIRCRAFT; DESIGNING, MANUFACTURING, ASSEMBLING, CLEANING, MAINTAINING OR REPAIRING AIRCRAFT, NOT OTHERWISE PROVIDED FOR; HANDLING, TRANSPORTING, TESTING OR INSPECTING AIRCRAFT COMPONENTS, NOT OTHERWISE PROVIDED FOR
    • B64F5/00—Designing, manufacturing, assembling, cleaning, maintaining or repairing aircraft, not otherwise provided for; Handling, transporting, testing or inspecting aircraft components, not otherwise provided for
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B23/00—Testing or monitoring of control systems or parts thereof
    • G05B23/02—Electric testing or monitoring
    • G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
    • G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
    • G05B23/0224—Process history based detection method, e.g. whereby history implies the availability of large amounts of data
    • G05B23/024—Quantitative history assessment, e.g. mathematical relationships between available data; Functions therefor; Principal component analysis [PCA]; Partial least square [PLS]; Statistical classifiers, e.g. Bayesian networks, linear regression or correlation analysis; Neural networks
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/0499—Feedforward networks
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/082—Learning methods modifying the architecture, e.g. adding, deleting or silencing nodes or connections
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/09—Supervised learning

Definitions

  • the present invention relates in general to a method for performing structural diagnostics, and more specifically to a method for performing diagnostics of a mechanical structure, in particular an aircraft structure, adapted to evaluate or monitor the presence of damage or defects caused in a structure by operating loads and/or events occurring while in service.
  • a method for predicting the behaviour of a structure subject to loads was developed by the same Applicant and described in the European patent application EP 2,281,224 A1.
  • the method comprises the provision of a mathematical model of the structure, detection of the state (deformation) of the structure in a plurality of primary points and in a plurality of additional points, determination of the loads acting on the structure and associated with the state detected in the primary points on the basis of the aforementioned mathematical model, estimation, using the loads determined, of the state of the structure in the additional points, and comparison between the state of the structure estimated and that detected in the additional points, so that an intact state of the structure is determined if the estimated and detected values of the state parameter match, or a defective state of the structure if these values differ.
  • the object of the present invention is to provide an improved method for performing structural diagnostics, which is both simple and flexible and allows the physical and mechanical conditions of a structure to be estimated with continuity in a reliable manner.
  • a further object of the invention is to provide a method for performing diagnostics which can be applied without the need for excessive calculation and in particular without the need to create a physical/mathematical model of the structure and which can therefore be implemented on-board an aircraft also when in service or during a mission.
  • the invention also relates to a system and a computer program for performing the diagnostics of a mechanical structure.
  • the present invention is based on the characterization of a mechanical structure being examined which is subject to operating loads able to cause local deformation thereof (or modify another parameter indicative of its state) and on the correlation in real time of real deformation data (or more generally data indicative of the real variation of the prechosen state parameter) and presumed deformation data (or more generally presumed variation of the prechosen state parameter), a comparison thereof being used to deduce the intact or defective condition of the structure.
  • a defect of the structure may consist of a hole, a filled hole or other modifications to the surface or volume, for example caused by the insertion of a connection member, impact damage, delamination, porosity, or due to a zone of the structure which has a different resin or fibre intensity.
  • a defect may be concentrated in a point with specific coordinates or spread out in a direction or over an area or within a volume of the structure.
  • the structure being examined is equipped with a limited number of deformation sensors located in relevant points.
  • a possible concentrated defect located far from them, may not cause any variation in the state of the structure at the relevant points, so that a given load or load vector gives rise to a deformation vector which is unchanged in the presence of a defect.
  • the criterion for choosing the detection points must preferably take into account the sensitivity to the structural defect at said points.
  • a neural network the degree of complexity of which depends on the morphological complexity of the structure, is trained on the basis of the state conditions detected on the structure at the relevant points by means of association with at least one and preferably a plurality of different load conditions.
  • the neural network is designed to estimate a correlation between the state or the variations in state detected in a subset of relevant points and the state or the variation in state in one or more residual relevant points.
  • a presumed variation in state of the structure under examination at a relevant point depending on a given operating load is estimated by means of the neural network which has been suitably trained and is compared with the corresponding assumed real value of the state parameter measured by the sensor associated with the relevant point.
  • an associative prediction of the state or modification in the state is assigned to a subset of relevant points and preferably to each relevant point of the complete set of relevant points on the basis of the state or modifications in the state detected at the other points of the set. Therefore, for each point and any load situation a comparison may be performed between the value of the state parameter predicted by the neural network for that point and the real value of the state parameter detected by the associated sensor, basically performing a comparison between the expected state of the structure and the detected state.
  • the diagnostic evaluation of the structure is performed by means of identification and signalling of the points where the value assumed by the state parameter detected differs from the expected value by an amount greater than a predetermined tolerance threshold.
  • An intact state of the structure is determined if the expected and detected values of the state parameter match within the predetermined tolerance threshold, or a defective state of the structure is determined if these values differ beyond the predetermined tolerance threshold.
  • the diagnostic evaluation may be conveniently verified by considering a plurality of different load situations and therefore measurements of the presumed variation of the state parameter, so that the existence of a mismatch condition between values predicted by means of the neural network and values detected by the sensors in a plurality of load situations may be interpreted as a confirmation of the presence of damage or a defect in the structure, while the existence of a mismatch condition between values predicted by means of the neural network and values detected by the sensors in a single load situation or in a small number of load situations together with the existence of a match condition between values predicted by means of the neural network and values detected by the sensors in a multiplicity of different load situations may be interpreted as an occasional signal.
  • the method according to the invention does not require the construction of a complex model of the diagnostics structure, for example finite-elements model, as described in EP 2,281,224 A1.
  • FIG. 1 shows an example of a diagnostics system according to the invention, applied to an aircraft
  • FIG. 2 shows an example of structure and a system of forces acting thereon
  • FIG. 4 is a diagram illustrating an example of a neural network according to the invention.
  • FIG. 1 shows the aircraft, generally denoted with A, and some of its structural parts which are to be monitored with regard to their intact or defective condition, for example the fuselage S 1 , the wing structure S 2 and the tail unit S 3 .
  • a plurality of sensors, generally denoted with P, are shown located on each part in N relevant detection points suitable for detecting a parameter indicative of the state of the aircraft structures, for example in the description provided here the local static deformation (where applicable, in more than one direction).
  • the sensors are connected to an electronic processing unit U to which respective signals representing the parameters detected are transmitted.
  • a database DB is associated with the processing unit and is designed to store a plurality of vectors comprising the values assumed by at least one predetermined state parameter detected at the N points in different load conditions.
  • a neural network is designed to determine a correlation between the values assumed by the state parameter in the N ⁇ 1 points different from the point P, and the value assumed by the state parameter in the point P i , depending on at least one and preferably a plurality of load conditions.
  • Each neural network is a network with Q levels, with d Q nodes per level, as shown in FIG. 4 .
  • Q levels with Q nodes per level, as shown in FIG. 4 .
  • a correlation of the neural type established between N relevant points X 1 , X 2 , . . . , X N at the input and a relevant point X F at the output is described.
  • a neural box consisting of Q successive lines (for example 3), each with dimensions d 1 , d 2 , . . . , d Q (for example 3 nodes per line) is established.
  • the correlative logic flow is shown in the figure, so that each node contributes to all the nodes of the next level.
  • a respective correlation parameter C is defined for each relevant input point for each neuron (inner node) and for each relevant output point.
  • the processing unit is able to provide an associative prediction of the value of the state parameter for the remaining point.
  • the processing unit is connected moreover to a signalling unit D for indicating to an operator, such as the aircraft pilot or a maintenance engineer, visually by means of written information and mapped points on a screen or electronically by means of issuing of a report, the intact or defective state of the monitored structures.
  • a signalling unit D for indicating to an operator, such as the aircraft pilot or a maintenance engineer, visually by means of written information and mapped points on a screen or electronically by means of issuing of a report, the intact or defective state of the monitored structures.
  • FIG. 2 An example of a structure being examined by a diagnostics system is shown in FIG. 2 , in the form of a fuselage panel of an aircraft—generally denoted with 10 and shown in a top plan view and side view—which comprises a flat bottom element 20 which has on a surface 22 a series of reinforcing ribs 24 .
  • L 1 -L K indicate the vectors representing the forces acting on the structure (which is essentially two-dimensional) in a predetermined operating condition, by way of example and for the sake of simplicity having components only in the plane in which the structure lies.
  • P i denotes relevant points on the surface of the structure, which are typically chosen based on a criterion of substantial periodicity, except for any clustering in the vicinity of areas which are more critical from a structural point of view (for example, the skin/reinforcement bonding zone, in order to diagnose any possible detachment of the reinforcements).
  • Deformation sensors of the type known per se are located in the N detection points (or relevant points of the structure) P i ; these sensors may consist, for example, of surface sensors or sensors which are embedded in the structure and which are connected (electrically, optically or wirelessly) to the processing unit of the diagnostics system on-board the aircraft designed to associate the signals acquired by the sensors with deformation values of the structure.
  • the diagnostics method according to the invention is described in detail with reference to the flow diagram shown in FIG. 3 .
  • the diagnostics method is implemented by the on-board processing unit U designed to execute groups or modules of processing and calculation programs stored on a disk or accessible on the network, for performing the procedures described.
  • step 100 the location of the relevant points on the structure is determined and the structure state sensors are positioned at these points.
  • the sensors may be located on the structure following determination of the topology of relevant points, or vice versa, using a network of pre-existing sensors on the structure a subset (or even the entire set) of corresponding relevant points is identified on the structure.
  • M state vectors V Sj [S 1j , S 2j , . . . , S Nj ] are acquired in step 200 for N relevant points and M different load conditions, with 1 ⁇ j ⁇ M, which are stored in the database DB.
  • each element of V S may assume a finite discrete number of values, for example owing to measurement discretization of the sensors which are employed on the structure.
  • the processing unit U acquires N training deformation values [ ⁇ 1l , ⁇ 2l , ⁇ Nl ], one for each relevant point P i .
  • the processing unit therefore acquires M deformation vectors, each of N points.
  • the M vectors of N points are stored in the database DB.
  • step 300 a step for training the N neural networks is performed (one for each relevant point), setting for the neural network associated with the i-th point P i a condition of input values equal to the values of the deformation detected in the N ⁇ 1 relevant points different from P i and stored in DB, and an output value representing the value of the deformation detected in the i-th relevant point P i , which is also stored in DB.
  • Each neural network creates an association between the deformations in N ⁇ 1 points and that in the relevant point P i with which it is associated, so that the processing unit has at its disposal N associative laws, of the type described above, for the value of the deformation of a point P i and each of the other N ⁇ 1 points, for each value of i lying between 1 and N.
  • Each neural network is configured during a training step advantageously performed during the first operating step of the structure.
  • the data of M different load conditions are used, where M may be chosen depending on the number of coefficients C used by the neural network and should conveniently be at least five times the number of coefficients C in order to achieve satisfactory training.
  • the relevant points are selected based on structural and statistical (variability) criteria.
  • operation of the neural networks may be verified in step 400 by comparing the output values envisaged by the trained network for given input values with the output values used during training, and assessing whether the difference, considered at a specific point and as an average, exceeds a fixed threshold and, in the case where incorrect operation is established (i.e. the difference exceeds, at a specific point and/or as an average, the fixed threshold of at least one or a predetermined minimum number thereof) the number of different load conditions to be used for performing detection of the state conditions in the relevant points is increased, generating new training deformation vectors [ ⁇ 1l , ⁇ 2l , . . . , ⁇ Nl ] which are stored in the database DB (step 200 ) and on which training of the neural networks ( 300 ) is carried out again.
  • the topology of relevant points is modified (step 100 ), by means of the addition or replacement of points, and then the steps for acquiring M training deformation vectors [ ⁇ 1l , ⁇ 2l , . . . , ⁇ N′l ] for N′ relevant points and M′ different load conditions, storage thereof in the database DB and training of the N′ neural networks are repeated in step 300 .
  • incorrect operation persists, in addition to prolonging the training period, it is possible to envisage modifying the number of levels and/or nodes per level of the neural network and/or modifying the function (function type) f and ⁇ .
  • the processing unit is configured to perform diagnostics of the structure, subject to any (for example periodic) updating of the deformation vectors, and corresponding new training of the neural networks, for example following modifications to the structure or ageing thereof.
  • ( ⁇ P ) q indicates a distribution of deformations in the grid induced by the same load or plurality of loads in the presence of a structural defect
  • ( ⁇ P ) d indicates a distribution of deformations detected by the sensors.
  • step 500 the current state ( ⁇ P ) d of the structure at the relevant points for a given current load condition is detected, for example the current deformations vector [ ⁇ 1d , . . . ⁇ id , . . . , ⁇ Nd ] is detected.
  • step 600 for each point P i with 1 ⁇ i ⁇ N the value of the deformation ⁇ ′ i is calculated by means of the associated neural network previously trained, using inputs including the deformation values detected in the other points ( ⁇ 1d , . . . ⁇ (i ⁇ 1)d , ⁇ (i+1)d , . . . , ⁇ Nd ).
  • step 700 for each point and any load situation a comparison is carried out between the value of the state parameter predicted by the neural network and the value of the state parameter detected by the sensor. Specifically, the comparison between the value of the deformation ⁇ id detected at the point P i and the value of the deformation ⁇ ′ i calculated by the respective neural network in the same point is performed, repeating the comparison for each i, where 1 ⁇ i ⁇ N.
  • An effective diagnostics evaluation is therefore performed by means of the comparison, at each point, between the expected structure state and the detected structure state. Identification of defects in the structure is performed for those points where the detected structure state differs from the expected state calculated by means of the respective neural network (namely there is a mismatch) beyond a predetermined percentage threshold.
  • the diagnostics method concludes that the structure is intact ( 800 ), signalling this condition by means of a signalling unit D to an operator, such as the aircraft pilot or a maintenance engineer, visually by means of written information and mapped points on a screen or electronically by means of issuing of a report, so as to indicate the intact state of the structure monitored.
  • the diagnostics method interprets a possible defective condition of the structure ( 900 ). Consequently, the method repeats the step 500 for detecting the state of the structure in the relevant points selected, during a successive instant, for the current load condition. It then repeats the step 600 for each point P i , where 1 ⁇ i ⁇ N, calculating the value of the state parameter by means of the associated neural network previously trained and, finally, again in step 700 , the comparison between the value of the state parameter detected at the point P 1 and the value of the state parameter calculated by the neural network at the same point, for each i, where 1 ⁇ i ⁇ N, is performed.
  • the cycle of operations in steps 500 - 700 is repeated a predetermined number of times, checking whether a predetermined number of repetitions in the comparison step 1000 have been reached, unless a condition where there is substantial match between the values, and therefore an intact condition of the structure, is definitively recognized.
  • a signal ( 1100 ) is emitted, by means of the signalling unit D, to an operator, such as the aircraft pilot or a maintenance engineer, visually in the form of written information and mapped points on a screen or electronically by means of issuing of a report, so as to indicate the defective state of the monitored structure and its location (namely, identification of the point P i where there is no match between the value of the state parameter detected and the value of this parameter calculated by the neural network).
  • the diagnostics evaluation may be further verified by considering different load situations and therefore measurements of the structure state: if the mismatch is repeated for different load conditions, this may be interpreted as a confirmation of the presence of damage or a defect in the structure which induces a variation in state. If the mismatch is not repeated, this may be interpreted as being an occasional or spurious signal, not caused by real physical factors.
  • mapping of the points where the presence of damage or a defect in the structure is determined provides an indication of the extent of the damage. For example, determination of damage or a defect in various adjacent points is an indication of a delaminated area.
  • the method concluded as illustrated in the flow diagram shown in the figure may be cyclically repeated, for example at predetermined periodic intervals in accordance with a predetermined monitoring program.
  • a surplus is created by increasing the number of relevant detection points so as to have a certain number of additional backup sensors.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Mathematical Physics (AREA)
  • Artificial Intelligence (AREA)
  • General Physics & Mathematics (AREA)
  • Computing Systems (AREA)
  • Software Systems (AREA)
  • Computational Linguistics (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Biophysics (AREA)
  • General Engineering & Computer Science (AREA)
  • Biomedical Technology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Health & Medical Sciences (AREA)
  • Automation & Control Theory (AREA)
  • Manufacturing & Machinery (AREA)
  • Transportation (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
  • Testing Or Calibration Of Command Recording Devices (AREA)
US13/933,964 2012-07-04 2013-07-02 Method for performing diagnostics of a structure subject to loads and system for implementing said method Active 2035-06-02 US9969507B2 (en)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
ITTO2012A0588 2012-07-04
IT000588A ITTO20120588A1 (it) 2012-07-04 2012-07-04 Procedimento per la diagnostica di una struttura sottoposta a carichi e sistema per l'attuazione di detto procedimento
ITTO2012A000588 2012-07-04

Publications (2)

Publication Number Publication Date
US20140012461A1 US20140012461A1 (en) 2014-01-09
US9969507B2 true US9969507B2 (en) 2018-05-15

Family

ID=46758936

Family Applications (1)

Application Number Title Priority Date Filing Date
US13/933,964 Active 2035-06-02 US9969507B2 (en) 2012-07-04 2013-07-02 Method for performing diagnostics of a structure subject to loads and system for implementing said method

Country Status (5)

Country Link
US (1) US9969507B2 (it)
EP (1) EP2682836B1 (it)
ES (1) ES2663251T3 (it)
IT (1) ITTO20120588A1 (it)
RU (1) RU2013130664A (it)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20220205870A1 (en) * 2019-09-30 2022-06-30 Airbus Operations Limited Test apparatus

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105527955B (zh) * 2014-09-28 2018-02-27 中国航空工业集团公司西安飞机设计研究所 一种飞机质量特性建模方法
CN104803009A (zh) * 2015-04-27 2015-07-29 中国航空工业集团公司沈阳飞机设计研究所 一种无人机地面综合检测系统及方法
RU2595066C1 (ru) * 2015-06-24 2016-08-20 Открытое акционерное общество "Лётно-исследовательский институт имени М.М. Громова" Способ оценки нагружения конструкции самолёта при лётных прочностных исследованиях с использованием искусственных нейронных сетей
RU2614740C1 (ru) * 2015-11-16 2017-03-29 федеральное государственное бюджетное образовательное учреждение высшего образования "Нижегородский государственный технический университет им. Р.Е. Алексеева" Способ оценки основного параметра, определяющего уровень и характер нагрузки при диагностике особо ответственных узлов транспортных средств
US20170283085A1 (en) * 2016-04-04 2017-10-05 The Boeing Company On-board structural load assessment of an aircraft during flight events
IT201800006499A1 (it) 2018-06-20 2019-12-20 Procedimento per la diagnostica di una struttura sottoposta a carichi basato sulla misura di spostamenti, e sistema per l'attuazione di detto procedimento.
EP3639199A1 (de) * 2018-06-29 2020-04-22 Renumics GmbH Verfahren zum bewerten eines zustands eines dreidimensionalen prüfobjekts und entsprechendes bewertungssystem
CN109752196B (zh) * 2019-01-28 2019-11-22 吉林大学 一种基于bp神经网络控制的车辆侧风试验方法
US12258144B2 (en) * 2019-05-30 2025-03-25 University Of Washington Aircraft wing motion prediction systems and associated methods
CN116577410B (zh) * 2023-05-16 2025-08-26 中国人民解放军国防科技大学 一种复合材料圆柱壳结构分层损伤识别方法
CN120403933B (zh) * 2025-07-04 2025-10-14 中建三局集团有限公司 一种钢绞线体内预应力分布监测方法及系统

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5774376A (en) * 1995-08-07 1998-06-30 Manning; Raymund A. Structural health monitoring using active members and neural networks
US6480792B1 (en) * 1998-10-22 2002-11-12 British Aerospace Public Limited Company Fatigue monitoring systems and methods incorporating neural networks
US7286964B2 (en) * 2003-09-22 2007-10-23 Advanced Structure Monitoring, Inc. Methods for monitoring structural health conditions
WO2010064216A1 (en) 2008-12-05 2010-06-10 Alenia Aeronautica S.P.A. Procedure for the prognostic of a structure subject to loads
US7822697B2 (en) * 2006-09-29 2010-10-26 Globvision Inc. Method and apparatus for infrastructure health monitoring and analysis wherein anomalies are detected by comparing measured outputs to estimated/modeled outputs by using a delay
US20110231037A1 (en) * 2008-09-19 2011-09-22 Valorbec Societe En Commandite Hard-landing occurrence determination system and method for aircraft
US20110313726A1 (en) 2009-03-05 2011-12-22 Honeywell International Inc. Condition-based maintenance system for wind turbines
US20130238532A1 (en) * 2012-03-12 2013-09-12 The Boeing Company Method and Apparatus for Identifying Structural Deformation
US20160091388A1 (en) * 2013-05-06 2016-03-31 Vrije Universiteit Brussel Effective structural health monitoring

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5774376A (en) * 1995-08-07 1998-06-30 Manning; Raymund A. Structural health monitoring using active members and neural networks
US6480792B1 (en) * 1998-10-22 2002-11-12 British Aerospace Public Limited Company Fatigue monitoring systems and methods incorporating neural networks
US7286964B2 (en) * 2003-09-22 2007-10-23 Advanced Structure Monitoring, Inc. Methods for monitoring structural health conditions
US20070260425A1 (en) * 2003-09-22 2007-11-08 Advanced Monitoring Systems, Inc. Systems and methods of generating diagnostic images for structural health monitoring
US7596470B2 (en) * 2003-09-22 2009-09-29 Advanced Structure Monitoring, Inc. Systems and methods of prognosticating damage for structural health monitoring
US7822697B2 (en) * 2006-09-29 2010-10-26 Globvision Inc. Method and apparatus for infrastructure health monitoring and analysis wherein anomalies are detected by comparing measured outputs to estimated/modeled outputs by using a delay
US20110231037A1 (en) * 2008-09-19 2011-09-22 Valorbec Societe En Commandite Hard-landing occurrence determination system and method for aircraft
WO2010064216A1 (en) 2008-12-05 2010-06-10 Alenia Aeronautica S.P.A. Procedure for the prognostic of a structure subject to loads
US20110313726A1 (en) 2009-03-05 2011-12-22 Honeywell International Inc. Condition-based maintenance system for wind turbines
US20130238532A1 (en) * 2012-03-12 2013-09-12 The Boeing Company Method and Apparatus for Identifying Structural Deformation
US20160091388A1 (en) * 2013-05-06 2016-03-31 Vrije Universiteit Brussel Effective structural health monitoring

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
Italian Search Report for corresponding Italian Patent Application No. TO2012A000588 dated Mar. 21, 2013.
Jiang, X. "Recent Development in Structural Damage Diagnosis and Prognosis", Recent Patents on Engineering, Jun. 1, 2010, pp. 102-121.
Nakamura, M. et al. "A method for non-parametric damage detection through the use of neural networks", Earthquake Engineering & Structural Dynamics, vol. 27, No. 9, Sep. 1, 1998, pp. 997-1010.
Xu, B. et al. "Direct identification of structural parameters from dynamic responses with neural networks", Engineering Applications of Artificial Intelligence, Pineridge Press, Swansea, GB, vol. 17, No. 8, Dec. 1, 2004, pp. 931-943.

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20220205870A1 (en) * 2019-09-30 2022-06-30 Airbus Operations Limited Test apparatus

Also Published As

Publication number Publication date
RU2013130664A (ru) 2015-01-10
ITTO20120588A1 (it) 2014-01-05
US20140012461A1 (en) 2014-01-09
ES2663251T3 (es) 2018-04-11
EP2682836A3 (en) 2016-04-20
EP2682836A2 (en) 2014-01-08
EP2682836B1 (en) 2017-12-20

Similar Documents

Publication Publication Date Title
US9969507B2 (en) Method for performing diagnostics of a structure subject to loads and system for implementing said method
US10891406B2 (en) Prediction methods and systems for structural repair during heavy maintenance of aircraft
EP3454289B1 (en) Plant abnormality detection method and system
CN112785091B (zh) 一种对油田电潜泵进行故障预测与健康管理的方法
BR102016011297A2 (pt) sistema integrado e métodos para o gerenciamento e monitoramento de veículos
TWI683058B (zh) 故障機率評估系統
US8706447B2 (en) Procedure for the prognostic of a structure subject to loads
US20140058709A1 (en) Structural health management system and method based on combined physical and simulated data
CN108027611B (zh) 利用受专家意见监督的决策模式学习的用于机器维护的决策辅助系统和方法
KR20170126786A (ko) 음향 방출 시험을 위한 방법 및 장치
US11436485B2 (en) Method for performing diagnostics of a structure subject to loads based on the measurement of displacements and system for implementing said method
CN114357378A (zh) 一种基于疲劳断裂风险分析的概率单机结构健康监控方法
CN120373110A (zh) 网架钢结构节点失效扩散的预测方法
CN120409143B (zh) 制动器闸瓦虚实一致性评判方法
EP2956751B1 (en) Method and monitoring device for monitoring a structure
CN118296848A (zh) 一种基于系统可靠性和dbn的飞机结构健康评估方法
KR101887955B1 (ko) 항공기 수명관리를 위한 표준비행데이터 생성방법
CN104361404B (zh) 机载无线数据通讯中继单元剩余寿命预测方法与预测装置
Khan et al. Prognostics of crack propagation in structures using time delay neural network
Colombo et al. On statistical Multi-Objective optimization of sensor networks and optimal detector
Sbarufatti et al. Advanced stochastic FEM-based artificial neural network for crack damage detection
CN119150021A (zh) 螺栓故障诊断模型的构建方法、系统、设备和介质
Shao et al. A Fusion Prognostics Framework for Electronic System
Kim et al. Study on Instrument Fault Detection using OLM Techniques for PHM Application in NPPs

Legal Events

Date Code Title Description
AS Assignment

Owner name: ALENIA AERMACCHI S.P.A., ITALY

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:IANNONE, MICHELE;REEL/FRAME:030737/0032

Effective date: 20130604

STCF Information on status: patent grant

Free format text: PATENTED CASE

MAFP Maintenance fee payment

Free format text: PAYMENT OF MAINTENANCE FEE, 4TH YEAR, LARGE ENTITY (ORIGINAL EVENT CODE: M1551); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY

Year of fee payment: 4

FEPP Fee payment procedure

Free format text: MAINTENANCE FEE REMINDER MAILED (ORIGINAL EVENT CODE: REM.); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY